Deep Learning Model for Facial Emotion Recognition

摘要

Facial expressions are manifestations of nonverbal communication. Researchers have been largely dependent upon sentiment analysis relating to texts, to devise group of programs to foretell elections, evaluate economic indicators, etc. Nowadays, people who use social media platforms to share their experiences or express themselves, primarily make use of images and videos. The methods for classification of these facial expressions have been studied over the years. There is strong evidence for the universal facial expressions of six emotions which include: happy, sadness, anger, disgust, fear, and surprise. Emotion is applicable in many domains such as gaming, health care centers, and burglary detection system. Emotion detection comprises of three stages viz. face detection from the given image, extracting its features, and classification. The techniques involved in these three major processes and their sub-processes are reviewed in this paper. Based on this survey, a deep learning model for facial emotion recognition has been put forth in this paper.